Meta presented a prediction last year with the products of the Truceful IQ, which will reach $ 2 billion to 3 billion dollars in revenue in 2025, and between 460 billion dollars and 1.4 trillion dollars by 2035, According to the court documents that were canceled on Wednesday.
Documents, provided by lawyers for books authors The prosecution of the definition of what they claim is unauthorized training Among the company’s artificial intelligence on their business, it does not indicate what Meta considers exactly the “AI Mottical Product”. But it is generally knowledgeable that the technology giant earns money – and stands to earn more money – from obstetric artificial intelligence in a number of flavors.
Meta has Revenue sharing agreements With some companies that host the Llama Open group of models. The company recently The application programming interface launched To customize and evaluate Lama models. Meta AI, AI’s assistant in the company, may eventually display ads and submit a subscription option with additional features, CEO Mark Zuckerberg He said During the company’s first -quarter profit call.
Court documents also reveal that dead spend a huge amount on its AI products.
In 2024, the company’s “Genai” budget amounted to more than 900 million dollars, and this year, it may exceed one billion dollars, according to the documents. This does not include the infrastructure needed to operate and train artificial intelligence models. Dead Previously said It plans to spend $ 60 billion to 80 billion dollars on capital expenditures in 2025, especially on the wide new data centers.
These budgets may have been higher if it included deals to license the books of the authors who sue Meta. For example, Meta discussed in 2023 to spend more than 200 million dollars to obtain Llama training data, about $ 100 million of which would go towards books alone, according to the documents. But the company decided to follow other options: Electronic books pirated on a huge scale.
Meta Techcrunch spokesman sent the following statement:
TECHRUNCH event
Berkeley, California
|
June 5
“META has developed Amnesty International Transformative Models (Open) that works to operate innovation, productivity and amazing creativity of individuals and companies. The fair use of copyright -protected materials is vital for this. We disagree with (authors) assurances, and the full record tells a different story.
https://techcrunch.com/wp-content/uploads/2025/02/GettyImages-2195497483.jpg?resize=1200,800
Source link